PROJECT SUMMARY/ABSTRACT In sub-Saharan Africa, where the burden of HIV-related disease remains high, adult prevalence is 15% and many women are newly diagnosed and/or initiated to ART during pregnancy. Many countries, including South Africa have adopted World Health Organization guidelines for PMTCT which include initiation of lifetime therapy for pregnant women at diagnosis regardless of clinical stage or CD4 count (Option B+). While estimates of participation in ART during pregnancy are high in South Africa (over 90%), the benefits of lifetime therapy can only be achieved if women are retained in care and treatment and adherent to medication. The postpartum period is a poorly understood time of transition and potential risk of loss from HIV care. In one Johannesburg based study, by 6 months postpartum, loss to care was 38%. Changes in residence and care location can also complicate the characterization of care during this time. We propose to address this gap by leveraging data and infrastructure from an ongoing collaboration based at a well-established Health and Demographic Surveillance Site (HDSS) in Mpumalanga province, South Africa with high antenatal prevalence of HIV. The existing collaboration supports collection of medical record data from all clinics serving the HDSS area with clinical records linked to an annual HDSS census. Approximately 115,000 people are covered by the HDSS area. Census data includes all births and deaths in the area, as well as household SES, household composition, and in and out migration. Using the linked clinical and census data in conjunction with a population representative community survey, we will construct a population-based open cohort of women during the pre-natal and post-partum period that provides data at individual, household, community, and clinic levels to address 2 research aims: Aim 1) Characterize engagement in HIV care following childbirth. Using data on all HIV related clinical visits, including medication dispensation and viral suppression data, we will describe loss from treatment and care for up to three years following childbirth for all women who are pregnant or give birth in the study area from 1 August 2015 through 31 July 2017. Linkage to the annual HDSS census and use of records across all area clinics will allow us to identify those who change residence or location of care following birth. Aim 2) Identify subgroups of women more likely to disengage from care and modifiable factors with the greatest potential to improve linkage and retention. We will identify characteristics of women most likely to disengage from care, clinics and communities with the highest rates of loss from care. Very little research has been done in the era of Option B+ that follows postpartum women beyond 6 weeks. As a result, these findings will be particularly useful. The strong collaboration at a research site where there is a focus on intervention development to improve community well-being positions this research well to translate findings on promising strategies to public health action to improve health and prevent onward transition of HIV.